CLEI Electronic Journal

Image classification using deep and classical machine learning models on small datasets: a complete comparative

Gonzalo Miranda Cabrera, Clemente Rubio-Manzano — Mon, 29 Apr 2024 00:00:00 -0600

One of the most important challenges in the Machine and Deep Learning areas today is to build good models using small datasets, because sometimes it is not possible to have large ones. Several techniques have been proposed in the literature to address this challenge. This paper aims at studying the different available Deep Learning techniques and performing a thorough experimentation to analyze which technique or combination thereof improves the performance and effectiveness of the models. A complete comparison with classical Machine Learning techniques was carried out, to contrast the results obtained using both techniques when working with small datasets. Thirteen algorithms were implemented and trained using three different small datasets (MNIST, Fashion MNIST, and CIFAR-10). Each experiment was evaluated using a well-established set of metrics (Accuracy, Precision, Recall, F1, and the Matthews correlation coefficient). The experimentation allowed concluding that it is possible to find a technique or combination of them to mitigate a lack of data, but this depends on the nature of the dataset, the
amount of data, and the metrics used to evaluate them.

SPLINE OPTIMIZATION OF SOFT CONNECTIVES IN MACHINE LEARNING MODELS

Vyacheslav Kalnitsky, Valery Vilkov — Mon, 29 Apr 2024 00:00:00 -0600

In this study, the problem of limited accuracy of machine learning models using soft logical connectives is investigated. Such connectives have shown their effectiveness in models with fuzzy initial data. On the one hand, the fundamental disadvantage of soft connectives is their non-associativity. On the other hand, the disadvantages of the currently used soft connectives include the loss of monotonicity and the inability to control several factors simultaneously. All these problems have been solved by the authors. We have proposed an approximation of the signum function by a ?¹-smooth spline. The first part of the spline is responsible for the curvature of the connective at the diagonal and is adjustable. The second part of the spline is the solution to the optimization problem. We minimized the difference between the connective and the associative connective, the latter in our study was the minimum function. In the resulting solution, the rate of deviation reduction is the highest among known connectives. We have achieved not only a small deviation from associativity, but also the presence of a large domain of exact associativity. This area is up to a third of the volume of all triples of arguments. A comparative analysis of the currently used soft connectives with the constructed model was carried out. It was shown that the spline approximation is able to reduce the influence of all negative factors and is more flexible in setting. Moreover, the constructed spline model allows numerous modifications depending on the factor that requires the most attention for different tasks.

A study on the interaction of the elderly with digital games through usability and user experience evaluations

Mon, 29 Apr 2024 00:00:00 -0600

The growing number of the elderly and their potential use of technological products requires understanding the interaction between these users and technology to develop appropriate products that include them. Accordingly, this article focuses on studying the interaction between older people and digital games to provide strategies for digital game developers to create inclusive games for these users. Thus, usability and user experience evaluation sessions were performed to achieve these objectives, and the older generation's acceptance of technology during interaction with digital games was verified. The data collection methods were observation and the application of questionnaires, one before the tests to evaluate the profile of the public and another afterward to assess usability, experience and technology acceptance. The test was applied to 36 people aged 60 and over. The results showed that the elderly prefer mobile devices and use them more often. Despite their lack of experience with digital games, the test participants had a good experience, even with difficulties such as game time management and perception of screen elements.

Migration of legacy Java desktop applications to collaborative Web

Antonio Labian, Jesús D. García-Consuegra, Manuel Ortega — Mon, 29 Apr 2024 00:00:00 -0600

The migration of legacy applications to more advanced technology platforms has been a recurring theme in the academic and business sectors. This transformation is driven by the desire to overcome the limitations of traditional desktop applications and integrate advanced remote access and collaboration capabilities. Essential features not only for technology upgrade but also to enable organizations to take advantage of remote collaboration.
The COVID-19 pandemic has accelerated this need, forcing many organizations to adopt remote work and online collaboration modalities, forcing them to transform their information systems. There are various methodologies and techniques, such as conversion, redeployment and wrapping, that offer different approaches to facilitate this migration by minimizing the effort required during the process. This paper specifically studies the migration of Java Swing applications to web environments, particularly those that incorporate collaborative functionalities. Several solutions that follow different migration strategies have been evaluated, performing a more detailed analysis of the Vaadin platform as an efficient tool to perform this migration, with limited effort and special attention to the inclusion of new collaborative functionalities. Finally, a migration procedure is defined that has been validated through two case studies.

From Trends to Experiences: Co-Creation with Generative Artificial Intelligence in Developing Interactive Multimedia Applications

Alexander Rozo-Torres, Carlos J. Latorre-Rojas, Wilson Javier Sarmiento — Mon, 29 Apr 2024 00:00:00 -0600

The emergence of generative artificial intelligence (GenAI) tools has ushered in new possibilities for creating diverse content, encompassing text, images, sound, videos, 3D objects, and even code for programming languages. Within this dynamic landscape, both significant challenges and opportunities arise in the field of interactive multimedia application development. It is imperative that the evolving trends in the utilization of these tools metamorphose into experiences that not only reflect but also facilitate the replication of models, methods, and development approaches. This paper contributes to the academic discourse by presenting three distinct development experiences. Each experience represents the initial steps of an incremental exploration process, thus contributing to exploratory research in this domain. These experiences are presented as case studies, with the first one analyzed retrospectively, and the latter two deliberately designed to scrutinize the challenges and opportunities associated with integrating generative artificial intelligence into the software development process, playing a significant role in a co-creation process. The results presented are qualitative, with detailed independent analyses for each case study, offering a comprehensive description of the findings. Thus, this paper presents a set of recommendations facilitating the transfer, replication, and improvement of these experiences by other multimedia development labs or teams.

Enhancing the Adaptative Gamification Process Through User Behavior and Context

Mon, 29 Apr 2024 00:00:00 -0600

This article presents an innovative approach to developing a strategy for situational awareness in an adaptive gamification framework within the context of Collaborative location-based collecting systems (CLCS). The proposed approach involves incorporating five key factors that represent user behavior and context in the adaptive gamification process for CLCS. These factors include player preferences, player status, gamified activities, groupware activities, and project goal status. Each factor is crucial in describing various aspects of the framework and groupware interaction that players should be aware of during their game experience. By acquiring knowledge about these five axes, users can analyze and react appropriately to various situations. This modeling approach enables the early development of awareness and facilitates decision-making. Ultimately, this article serves as a useful guide for improving existing frameworks and enhancing the overall user experience through situational awareness.

Preface to the Iberoamerican Human-Computer Interaction Conference 2023 Special Issue

Valéria Farinazzo Martins, BRUNO DA SILVA RODRIGUES — Mon, 29 Apr 2024 00:00:00 -0600